210 DISCOVERY REPORTS 



surface, subsequently sinking and developing towards maturity as they went down. For in the 

 unlikely event of the sinking eggs becoming neutrally buoyant at say 1500 m., and continuing to 

 develop there without further sinking, we should have expected them to be, if in fact shed at this level 

 and captured immediately or very shortly after, largely unsegmented (p. 184) and not in the very 

 advanced state of development in which so many of them were found. It is obvious, of course, from 

 their condition that the majority must have been captured long after they were laid since the eggs 

 when first laid are unsegmented. If then they did not as suggested sink from a higher level, but 

 were in fact laid where captured, it must be supposed that they sank still farther, subsequently 

 being returned to the 1500 m. level, many of them now practically ready to hatch (p. 102), by 

 the marked upwelling of bottom water that had evidently taken place in this locality. 



Conclusions and future research 



Although certain facts seem clear, as for instance (i) the enormous scale of the spawning that appears 

 to be associated with the shelf and slope waters of the far western Weddell drift and, in a slightly 

 lesser degree in higher latitudes, with the shelf and slope waters of the continental land, (2) the 

 absence, or virtual absence, of both spawning and hatching on the South Georgia whaling grounds 

 and in the West Wind drift, and (3) the enormous scale of the deep hatching that occurs in the oceanic 

 water of the Weddell and East Wind zones, much obviously has yet to be done before the whole 

 complex problem of the spawning can finally be solved. Our failure, for example, to find satisfactory 

 evidence of large-scale spawning deep down in oceanic water, despite the scale of the hatching there, 

 our failure equally to find satisfactory evidence of large-scale hatching in shelf seas, and above all 

 the influence it seems the sinking shelf water and resultant bottom current may have on the liberated 

 eggs, are all matters that call for much further investigation. 



It seems clear that future work should include the following : 



(i) Systematic exploration from November to April, with vertical closing nets, of the cold bottom 

 water at great depths below the Weddell stream, special attention being given to depths between 

 3000, or even deeper, and 1500 m.^ This it seems certain would fill in the missing detail of the 

 commencement of the development ascent, bring to light the masses of hatching eggs that surely 

 must exist somewhere between these levels, reveal the true bathymetric range of the First Nauplius 

 and demonstrate whether or not its existence was ephemeral. 



(2) The repetition of these observations, with the same objects in view, in the oceanic water of the 

 East Wind drift. 



(3) Systematic exploration from November to April, with large towed closing nets, at great depths, 

 between 3000, or again deeper, and 1500 m., of the oceanic water of the Weddell and East Wind 

 zones. This, if done thoroughly enough, would no doubt demonstrate conclusively whether or not 

 the females go deep to spawn, and if they do so as scattered individuals or as swarms. 



(4) Systematic exploration from November to April, with both vertical and towed nets, in the shelf 

 and slope waters of the western Weddell drift and of the Antarctic continental land at all levels from 

 the surface to the bottom, particular attention being paid to the sea bed where it seems the eggs might 

 be laid and where, in slope waters at least, they would certainly appear (p. 202) to be hatching. 



^ To save precious ship's time, and achieve rapid serial sampling at these great depths over as wide an area as possible, 

 a prime essential will be the development of a multiple-sampling vertical net on the lines described by Motoda (1953). Using 

 such a net, with small terminal store-nets revolving into position, Motoda was able to obtain five separate, successive vertical 

 samples in a single haul. A multiple vertical plankton sampler, employing different mechanical principles, has more recently 

 been described by Be, Ewing and Linton (1959), and Motoda (1959) produces many other ideas. Indeed, as Sysoiev (1957) 

 observes, one of the major tasks confronting oceanographical instrument-makers today is the construction and perfecting of 

 apparatus for observations at great depths. 



